Datatron announced enhancements to its MLOps and AI governance solution, making it even easier for enterprises to catalogue, operationalise, monitor and govern AI/ML models.
With Datatron, customers experience 15 to 20 times more effectiveness in model deployment, bringing substantial business gains and productivity improvements. Datatron also eliminates the complexity and expense associated with constant iteration and management of many AI models at one time.
Key enhancements to the Datatron Reliable AI platform include:
- ML Gateways: ML Gateways provide centralisation and orchestration of models and data in complex, multi-tenant environments. It’s designed to support a growing number of use cases, helping enterprises overcome challenges, including compliance, differing model technologies, and AI ownership across subsidiaries, partners, and internal data science teams
- Customer-defined KPIs: This enables enterprises to define their own formulas for continuous analysis of statistics and measures, set thresholds for warning and alert conditions, and include KPIs in the central governance dashboard
- Explainability with confidence: This unique innovation is a departure from many theoretical exercises by others. Datatron builds in a confidence score that is used against explainability, helping customers understand what data was relevant in the results and the level of trust one can place in those results
- Native Jupyter support: Supports direct import of Jupyter notebooks by data scientists to silently run alongside current models to get faster validation of fit, making all the governance metrics available before the model goes live
- Rapid setup and deployment: A new five-step guided process allows customers to run a selected model in production as APIs for real-time inferencing or scheduled batches in less than 10 minutes
Also Read: Which Tech Trends Will Impact Your Business?
